No description available.

No description available.

No description available.

The T-cell lymphoma invasion and metastasis (Tiam) family of proteins are guanine nucleotide exchange factors (GEFs). The Tiam family has two homologs, Tiam1 and Tiam2, which activate the Rho-family GTPase Rac1 and are crucial for cell-cell adhesion. Tiam1/2 contain several protein-protein interaction domains, in particular a PDZ domain. PDZ domains are distributed from C. elegans to H. sapiens proteomes. The prototypical ligands of PDZ domains are specific sequences located at the C-termini of their binding partners. In this thesis I examined the specificity of wild-type (WT) and two mutant forms of the Tiam1/2 PDZ domains, which bind to three ligands derived from adhesion receptors - Syndecan1 (SDC1), Caspr4 and Neurexin1 (NRXN1). The specificity of the Tiam1 PDZ WT domain for individual SDC family members was identified and found to be determined by distinct electrostatic interactions. Based on X-ray crystallographic studies, two ligand binding pockets were crucial for SDC1 binding while an additional pocket was important for accommodating the phosphate group in phosphorylated SDC1 (pSDC1). Methyl relaxation experiments of PDZ/SDC1 and PDZ/pSDC1 complexes revealed that PDZ-phosphoryl interactions dampened dynamic motions in a distal region of the PDZ domain by decoupling them from the ligand-binding site. Our data suggest a selection model by which specificity and phosphorylation regulate PDZ/SDC interactions and signaling events. Previous studies showed that the Tiam1 and Tiam2 WT PDZ domains have distinct ligand specificities. Moreover, a mutant of the Tiam1 PDZ domain that has four non-conserved residues substituted to the analogous residues in the Tiam2 PDZ domain (designated QM) showed switched specificity. Here, I determined the crystal structures of the free and ligand-bound Tiam1 PDZ QM. These structures revealed that two enlarged ligand binding pockets and a favorable electrostatic interaction are critical for the changed specificity. Complementary NMR studies showed enhanced dynamic motions in the Tiam1 PDZ QM domain. These data support a conformational selection model where mutations in the PDZ domain lead to the specificity switch through structural changes in binding pockets and enhanced dynamic motions. The Tiam2 PDZ QM domain was produced by substituting the four specificity residues to those in the Tiam1 PDZ WT domain. The Tiam2 PDZ QM had a switched specificity, similar to that found in the Tiam1 PDZ WT. NMR-based HSQC spectra and quantitative relaxation studies revealed that the Tiam2 PDZ WT domain had dynamic motions in several loop regions, while motions in the QM PDZ domain were extended to additional regions. Thermodynamic analyses of Tiam2 PDZ mutated domains uncovered energetic cooperativity between two binding pockets, with respect to both ligand binding and protein folding. I propose that the altered dynamic and thermodynamic properties support the ligand binding specificity switch in the Tiam2 PDZ QM. The literature indicates that aberrant SDC1 and Tiam1-Rac1 signaling pathways contribute to tumor progression. These two pathways are coupled by the PDZ-mediated Tiam1/SDC1 interaction. Here, I integrated structure-based in silico docking and an unbiased high throughput screen to discover inhibitors targeting this PDZ-mediated interaction and in order to inhibit downstream Tiam1-Rac1 signaling and tumor progression. Several compounds were identified and validated to block this interaction and downstream Rac1 signaling. Further modification of these molecules could lead to inhibitors with more potency and thus therapeutic values. In summary, detailed biochemical, structural and dynamics studies were performed to understand the mechanistic origins of ligand specificity in the Tiam1 and Tiam2 PDZ domains. These results highlight the significant structure-dynamics-function relationship in the PDZ domain and imply that rewiring these interactions is possible. Moreover, a high throughput screen identified small molecule inhibitors targeting these PDZ domains. These inhibitors could be used as molecular probes in cells to dissect the physiological roles of signaling cascades controlled by Tiam GEFs.

Replication protein A (RPA) is the major eukaryotic single-strand DNA (ssDNA) binding protein. RPA is composed of three subunits, RPA1, RPA2 and RPA3. RPA is essential for replication, repair, recombination, and checkpoint activation, and is required for maintaining genome integrity. In the cell, RPA binds to ssDNA intermediates and ensures that the appropriate pathway correctly processes them. The ssDNA-binding activity of RPA is primarily mediated by two high-affinity domains in the RPA1 subunit. DNA binds to these domains by interacting with polar and aromatic residues in a DNA-binding cleft in each domain. The aromatic residues are highly conserved and when mutated cause a separation-of-function phenotype. Mutation of the conserved aromatic residues in the high-affinity binding domains of RPA only modestly affected the affinity of RPA but these aromatic residue mutants were unable to support DNA repair while functioning in DNA replication. To understand the molecular basis of this phenotype, I have characterized the interactions of the aromatic mutants with different length ssDNAs and partial duplex DNA structures like those found in DNA repair. I also probed the conformations and dynamics of RPA-DNA complexes. My studies identified that there are at least two kinetic states when RPA binds to ssDNA that differ in their rate of dissociation from the DNA. I also showed that the aromatic residues are required for the stable binding to short ssDNA and contribute to the formation of the more long-lived state. My studies also showed that the more stable state is important for RPA in melting secondary DNA structure. We conclude that melting activity and/or stable binding by RPA is required for DNA repair but dispensable for DNA replication. These studies enhance our understating of molecular interactions between RPA and DNA that contribute to different cellular functions. The kinetic states in RPA could reflect changes in domain interactions or changes in conformation of the RPA-DNA complex. To try to understand the molecular basis of the different kinetic states, I used single molecule FRET analysis to characterize the spatial location of RPA domains and conformational dynamics in RPA-DNA complex. My studies showed RPA binds different locations along ssDNA and that generally RPA does not undergo global changes in conformation when bound to ssDNA. However, with a subset of label locations, some RPA-DNA complexes showed rare changes in conformation. These observations were most consistent with partial microscopic dissociation (domains of RPA partially dissociate from DNA, but has not yet equilibrated with the surrounding solution) of domains of RPA near the 3’ end of the complex and interactions of the flexible N-terminal, regulatory domain of RPA with the free DNA. My data suggests that the microscopic dissociation can occur without affecting the global structure of the RPA-DNA complex. These studies illustrate that different DNA metabolic pathways require different types of RPA-DNA complexes and that high affinity binding is not sufficient for all RPA functions. Specifically, my studies showed that DNA repair pathways require different ssDNA interactions. This suggests that modulation of the binding of individual domains and/or inter-domain interactions regulates the properties of the RPA-DNA complex and, in turn, that this could direct ssDNA intermediates into different pathways for processing. Together, my studies highlight the importance of dynamics in RPA binding to properly maintain the integrity of the genome.

Cells are subject to a wide variety of forces throughout their lifetimes. During epithelial morphogenesis, epithelial cells form sheets of cells that line the cavities and surfaces of organs. These cells protect the organs and are responsible for sensing and responding to mechanical forces. The ability of cells to respond and adapt to forces underlies the etiology of a variety of diseases including cardiovascular disease, cancer, and lung dysfunctions. Therefore, it is critical to understand how force affects cells. Cells sense forces through cell surface adhesion receptors. These receptors trigger reinforcement of cell anchoring junctions to counter the applied force. Anchoring junctions occur in two functionally distinct forms. The first type, adherens junctions, mediate cell-cell adhesion while the second type, focal adhesions, bind cells to the extracellular matrix. Each anchoring junction contains specific transmembrane adhesion receptors. These receptors in adherens junctions are the Ca2+-dependent transmembrane adhesion proteins known as cadherins. In epithelia, the cadherins are denoted E-cadherin or epithelial cadherin. For the second type of cell anchoring junction, the transmembrane receptors are the integrins. In response to force, epithelial cells reinforce the cell anchoring junctions and the actin cytoskeleton. Force on E-cadherin causes clustering of E-cadherins and stimulates the recruitment of β-catenin, α-catenin, and actin, thereby triggering growth of the adhesion complex. Force also stimulates a signaling pathway involving RhoA and phosphorylation of myosin light chain culminating in more actin stress fibers and an overall increase in actin polymerization. More specifically, force on E-cadherin actives AMPK. Active AMPK stimulates Abl activity to phosphorylate vinculin Y822. This phosphorylation event promotes RhoA-mediated cytoskeletal rearrangements, a process known as cell stiffening. The ability of a cell to regulate its actin cytoskeleton is critical for maintaining the balance of force between itself and its surroundings. Disruptions in actin cytoskeletal rearrangements underlie the progression of many diseases, including cardiovascular disease, lung dysfunctions, and cancer. Therefore, elucidating the signaling pathway that culminates in changes to the actin cytoskeleton is critical to understand the basic mechanisms that go awry in these diseased states. The second chapter of my thesis focuses on identifying a novel component of E-cadherin force transmission. I focus on the apoptotic regulator p21-activated kinase 2 (PAK2) as previous studies have implicated PAK2 in cell-cell junction mechanics. To examine its role in force transmission, I applied a variety of types of mechanical force to epithelial cells and monitored PAK2 activity. I observed that force results in activation of PAK2 and triggers its recruitment to the cadherin adhesion complex at cell-cell junctions. Furthermore, PAK2 is required for cell stiffening. Loss of PAK2 results in decreased E-cadherin and F-actin enrichment as well as loss of critical phosphorylation events in the E-cadherin force transmission signaling cascade. Finally, I demonstrated that PAK2 also plays a critical role in determining cell survival in response to force. Under normal physiological levels of force, AMPK binds PAK2 and protects it from caspase-3 mediated cleavage. As the amplitude of force increases, AMPK no longer binds PAK2 and PAK2 is cleaved. The cleaved product translocates to the nucleus where it promotes transcription of pro-apoptotic genes. This work provides a paradigm for how different amplitudes of force affect cell survival and calls for more in depth consideration of force applications in future studies. The third chapter of my thesis focuses on identifying a negative regulator of E-cadherin force transmission. Evidence suggests that the protein tyrosine phosphatase, SHP-2, may be involved in E-cadherin negative regulation. SHP-2 localizes to cell-cell junctions, and cells expressing constitutively active SHP-2 have similar phenotypes to cells expressing a phospho-mutant Y822F vinculin. I demonstrate that vinculin Y822 phosphorylation is followed by SHP-2 phosphorylation and activation. Active SHP-2 de-phosphorylates vinculin Y822 directly, halting the cell stiffening response. This work is the first specific negative regulator of E-cadherin force transmission. Together, these findings reveal the importance of strict regulation of cell stiffening, and provide a framework for exploring how diseases involving contractile disturbances.

Recent studies have shown the importance of enzyme regulation through the formation of higher order structures. This is exemplified by Acetyl coenzyme A Carboxylase, and Cytidine Triphosphate synthase, two enzymes that form long chain filaments in the presence of specific activators [5-7,14,22,26]. Recently, yeast glucokinase (Glk1p) was found to form filaments in the presence of increasing concentrations of Glucose-6-phosphate (G6P), the product of the reaction that it catalyzes [18]. Due to this fact, and the consistent appearance of what seem to be higher order and oligomer like peaks in size exclusion chromatography (SEC) chromatograms of human glucokinase (GK) expressed from Escherichia coli, an investigation of the nature of these peaks was conducted. Through the use of SEC, a method described in Tayyab at al 1992 was used to determine the molecular weight [38]. This resulted in the discovery that only the oligomeric peak was formed over time, was the approximate size of a trimeric glucokinase structure. SEC also discovered that factors, including time and ligand presence, affected the formation of the oligomeric complex, while spectrophotometric assays of the oligomeric protein indicated a decrease in overall activity. The oligomeric form of GK produced little to no activity in the presence of increasing concentrations of its substrate, glucose. This finding is consistent with the fact that the oligomer is an inactive form of GK that is not affected by glucose concentrations. / A Thesis submitted to the Department of Chemistry and Biochemistry in partial fulfillment of the requirements for the degree of Master of Science. / Spring Semester 2019. / April 19, 2019. / glucokinase, structure / Includes bibliographical references. / Brian G. Miller, Professor Directing Thesis; Lei Zhu, Committee Member; Michael Roper, Committee Member; Scott Stagg, Committee Member.

A number of physiologic processes require the expression of smooth muscle alpha actin (SMαA) to mediate cellular contraction. Stable expression of SMαA in differentiated vascular smooth muscle cells is associated with a contractile phenotype that is essential for regulation of blood flow and pressure. The transient expression of SMαA in myofibroblasts during wound repair facilitates wound closure. Hence, it is no surprise that dysregulation of SMαA gene expression in both cell types can have pathological consequences. Indeed, aberrant SMαA gene regulation has been implicated in diseases such as atherosclerosis and fibrosis. Therefore, a better understanding of the molecular mechanisms that regulate SMαA gene expression is necessary to uncover the factors involved in modulating the phenotype of vascular smooth muscle cells and fibroblasts in diseases affecting blood vessels and connective tissue. Previous studies have shown that the SMαA gene is regulated by a combination of transcriptional activators, repressors, and cofactors. Two members of the purine-rich element binding protein family known as Purα and Purβ have been implicated in the repression of SMαA gene expression. Both proteins bind to single-stranded, purine-rich DNA sequences in the SMαA gene promoter with high affinity and specificity. However, published loss-of-function and gain-of-function analyses suggest that Purβ is the dominant repressor in vascular smooth muscle cells and fibroblasts. Thus, the principal objective of this dissertation project was to define the specific molecular mechanism(s) by which Purβ represses the SMαA gene. This undertaking was made possible by the prior identification of a functional core region in the center of the protein containing three regions of internal homology termed repeats I, II, and III. Amino terminal repeats I and II form an intramolecular DNA-binding domain, while two carboxy terminal repeats III form an intermolecular dimerization domain. Further analysis revealed that the dimerization domain is also capable of interacting with purine-rich single-stranded DNA and is absolutely necessary for the full SMαA gene repressor activity of Purβ. In this dissertation, experimental findings are presented indicating that Purβ binding to single-stranded DNA is mediated by both ionic and hydrophobic interactions. Site-directed mutation of specific positively-charged amino acid residues in each of the three repeats resulted in reduced repression of the SMαA gene promoter owing to diminished DNA binding affinity. Mutation of a positionally-conserved arginine residue in the third repeat had the most significant effect on the function of Purβ. In addition, biochemical characterization of rare single nucleotide polymorphism-encoded variants of Purβ revealed that other amino acid changes in the third repeat affect protein-protein interaction, but not DNA-binding activity. Lastly, evidence is shown indicating that Purβ inhibits the potent smooth muscle-restricted co-activator myocardin via a novel protein-protein interaction based mechanism. Interestingly, specific point mutations and variations in the third repeat impair the ability of Purβ to repress myocardin cofactor function. Collectively, these studies demonstrate that the third repeat/dimerization domain of Purβ is essential for full repression of the SMαA gene as specific amino acid residues in this region mediate both protein-DNA and protein-protein interactions.

Cells synthesize proteins, the molecular instruments of all cellular processes, via intermediate biomolecules that are susceptible to damage at every step. Known as the central dogma of molecular biology, genes encoded in deoxyribonucleic acid (DNA) are transcribed, spliced, and matured into messenger ribonucleic acid (mRNA). These nucleic acids direct protein synthesis by the pairing of nucleotide triplets with transfer RNA (tRNA). tRNAs concomitantly decode the so-called codon, as they escort the correct amino acid to the ribosome for extension of the nascent polypeptide chain. Damage to any of these intermediate biomolecules can be highly damaging to protein synthesis, leading to aberrant biochemical processes, aging, cancer, or apoptosis. Accordingly, cells have evolved essential response and repair pathways to ensure that replication, transcription, and translation occur with high fidelity. In this dissertation, we interrogate two enzymes involved in these quality-control measures: 1) a DNA glycosylase which recognizes damage to the DNA bases, and 2) a tRNAHis guanylyltransferase-like protein (or THG1-like proteins, TLPs) which repairs truncated or mismatched tRNA via 3’5’ polymerization. DNA is assaulted daily to the tune of 30,000 lesions per cell per day by exogenous and endogenous stressors. One of many DNA repair pathways, the base excision repair (BER) pathway, removes the small non-bulky, and oxidized DNA lesions from the genome. DNA glycosylases are the first enzymes in the concerted mechanism tasked with scanning the entire genome for DNA damage and initiating the repair of lesions. The human genome encodes 11 DNA glycosylases, which possess overlapping substrate specificities within BER. The DNA glycosylase, endonuclease three (Nth), recognizes and removes oxidized pyrimidines during all phases of the cell cycle. We have solved the first X-ray crystal structure of human Nth-Like 1 (hNTHL1), which revealed a novel open conformation. This unprecedented example of an Nth DNA glycosylase undergoing interdomain rearrangement provides important insight into the molecular mechanism of this critical guardian of the genome. In eukaryotes, tRNAs must be modified at the 5’ end during maturation. tRNAHis guanylyltransferase (THG1), an essential gene in yeast, catalyzes the addition of guanine to the 5’ end of tRNAHis. Reverse polymerization requires adenylation (or guanylation) to activate the 5’ end of the tRNA. After adenylation, there is a shift of the 5’-phosphate of the tRNA to accommodate the forthcoming nucleophilic attack by the 3’-OH of the incoming nucleotide. In contrast to their human counterparts, the archaeal TLP enzymes utilize the 3’ to 5’ NTP-polymerization reaction to repair 5’-degraded tRNA molecules. We have solved the first crystal structure of a TLP caught in an intermediate step following activation by guanylation, showing that the base rotates within the nucleotide binding site to align the active site.

Determining the fragility of the red blood cell (RBC) is important for the diagnosis of and evaluation for treatment of several RBC diseases. In part RBC production is controlled through the hormone erythropoietin secreted by the kidney. In a previous study from this laboratory, it was found that RBC were more stable in uremic male rats compared to controls. In this experiment, uremia was induced in four groups of female rats through a two stage nephrectomy. The nephrectomy involved the removal of two-thirds of the left kidney, followed by the removal of the entire right kidney one week later. The animals were divided into three groups; NX-(5/6 nephrectomy), SH-(sham surgery), and PF-(sham surgery, but were fed the same food weight as the NX animals). The samples obtained in Trial I and Trial II were divided into two categories; initial and final. The initial samples were collected 14 days after the sham and five- sixth nephrectomy surgeries. The final sample were collected at the time of sacrifice. The samples obtained in Trial 11 consisted only of initial samples, taken fourteen days after the five-sixth nephrectomy and sham surgeries were completed. The samples in Trial IV were final samples, obtained at the time of sacrifice. Decreasing hypotonic %NaCl solutions were used to determine the hemolysis of RBC from the rats. RBC hemolysis was determined spectrophotometrically by monitoring hemoglobin absorbance at 540nm in the supernatant fluid. Analytic precision experiments using multiple assays of the same blood sample for 50% RBC hemolysis showed a coefficient of variation of only 1.1%. Analysis of the %NaCl at 50% RBC hemolysis did not differ significantly between the three groups of animals suggesting that although the NX animals were uremic, the RBC did not differ in stability to hypoosmotic shock. Future direction for this type of research will be extended to human studies where kidney failure patients (dialysis patients) can have both the age of the RBC and their fragility determined under therapy. The erythrocyte hemolysis peroxide test (HPT) was also performed on rat blood samples from Trial IV and on five human blood samples, in order to determine hemolysis in the RBC. A 2% H2O2 solution was used to determine RBC stability and %Hemolysis was calculated by dividing the value for hemolysis due to H2O2 by the 100% hemolysis value and multiplying by 100. Analysis of the %Hemolysis by HPT for each animal sample also showed no significant difference between the three subgroups of animals. Future direction for this type of research will be extended to include all four trials of animals as well as human studies involving patients with kidney failure (dialysis patients).